‘Screening Smarter’ for Prostate Cancer

March 23, 2018

During the NCCN Annual Conference in Orlando, Dr. Peter Carroll from UCSF discussed the evolution of screening tests for prostate cancer.

Screening for prostate cancer with the prostate-specific antigen (PSA) test, once widely promoted for all men 50 years and older, has fallen out of favor as a population-based screening tool. But when it comes to PSA testing, the proverbial baby shouldn’t be thrown out with the bathwater, experts caution.

“The right solution was not to ‘screen none’ but rather to screen smarter,” said Peter R. Carroll, MD, MPH, from UCSF Helen Diller Family Comprehensive Cancer Center in San Francisco, during the National Comprehensive Cancer Network (NCCN) Annual Conference, held March 22–24 in Orlando, Florida.

Carroll explained that during the 1990s and 2000s, PSA screening was poorly implemented; older men were overscreened, younger men were underscreened, low‐risk disease was overtreated, and high‐risk disease was undertreated.
“[D]espite these problems, we drove down prostate cancer mortality rates by more than 50%, but at the cost of too much entirely avoidable treatment and its attendant side effects,” he said.

Last year the United States Preventive Services Task Force (USPSTF) revised it recommendations on PSA screening and changed its “D” recommendation against PSA testing for men ages 55 to 69 and replaced it with a “C” recommendation. The task force now says that the men in that age group should decide individually with their doctors whether and when to undergo PSA testing.

The USPSTF changed their recommendations for the early detection of prostate cancer based largely on two issues; randomized trials show a mortality benefit from screening, and overtreatment has been tempered by increased utilization of active surveillance.

So where does that leave PSA testing now?

Carroll pointed out that screening recommendations from the different professional organizations vary but NCCN believes in baseline screening, which is different from the guidelines of other groups. While some of the details vary, all of the groups recommend shared decision making and have moved away from broad population-based criteria.

One of the previous problems with PSA testing was that it was a low-value test. “We were testing outside of the target population and testing too frequently,” he said, and the low specificity of the PSA test was also leading to frequent and unnecessary biopsies.

The detection/treat paradigm is now changing from a model of “detect all, treat all” to “detect some, treat some.” The new approach is to have a target population (age, risk factors, etc), to stop screening after low PSA results/later age, and to increase screening intervals. Also important is knowing what to do if a patient has a high PSA level, he explained.

More stringent indications for biopsy are needed, Carroll said. The 2018 NCCN guidelines for early prostate cancer detection call for a repeat PSA and a digital rectal exam if one had not previously been performed during the initial assessment. The next step would be a work-up for benign disease.

Before going to biopsy, clinicians should consider testing for biomarkers which will improve screening specificity for high-grade disease or using multiparametric MRI, he emphasized. New tests for specificity include the 4Kscore, which tests for PSA, free PSA (fPSA), intact PSA, and kallikrein-related peptidase 2; the ExoDx Prostate (IntelliScore), which tests for SPDEF, ERG, and PCA3; the Michigan Prostate Score (MiPS), which tests for PSA and the two genes PCA3 and TMPRSS2:ERG; and the Prostate Health Index, which tests for PSA, fPSA, and [‐2]proPSA.

Research has demonstrated that these tools can help men to avoid unnecessary biopsies; for example, the assessment with the 4Kscore Test avoided 36.2% of biopsies in one study and 51.3% in another.

“Prostate cancer early detection in well-informed healthy men saves lives,” Carroll emphasized, while cautioning that the optimal screening regimen for men at greatest risk (those who are African American, have a strong family history of prostate cancer, and have high‐risk germline mutations) is still not known.

Carroll added that that tests of specificity should be considered in men who wish to avoid a biopsy, and that “active surveillance is the preferred option for men with low-risk and very-low-risk disease.”